The goal of this work is to learn where and when neuroepithelial cells in the brain make the decisions that determine the type of neuron or glial cell they eventually become. In particular, we want to know the extent to which intrinsic (lineage-derived) and extrinsic (environmental) factors influence phenotypic choice. Our experimental preparation for this study is the chick optic tectum, chosen because it is experimentally accessible and well-studied, and because it shares features of radial, laminar, and topographic organization with the less accessible mammalian cerebral cortex. Our major experimental method will be a technique of retrovirus- mediated gene transfer that we developed a few years ago. In this method, we infect a progenitor cell with a retrovirus in vivo, then use a histochemical stain for the retroviral gene product to identify the descendants of the infected cell. Using this method, we will first document the migratory paths of clonally-related cells. Our preliminary studies show that clonal cohorts begin to ascend from the ventricular surface along a restricted radial path, but then diverge into three streams: most of the cells continue radially, but small subsets follow two distinct tangential paths and acquire specific phenotypes. This pattern suggests a relationship between migratory path and cell fate. Accordingly, we will combine retroviral labeling with other methods to map the three streams, identify structures that act as migratory guides, and seek adhesive molecules that might influence migratory choices. Then, in a third set of experiments, we will construct and use retroviral vectors that transfer two genes--the marker plus a second, putatively neuroactive gene or its antisense copy. By producing and analyzing small cones of "transgenic" cells in an otherwise unperturbed environment, we hope to uncover some mechanisms that regulate neural differentiation and migration. For example, we will be able to ask what roles particular adhesive molecules play in migratory choices, and whether altering a cell's migratory path alters the phenotype it adopts. Finally, as time permits, we will apply these approaches to other areas--e.g., forebrain and cerebellum--that differ from each other and from tectum in their organization. These comparisons will provide insight into the variety of genealogical and migratory strategies that the nervous system uses to generate diversity.